Skip to contentSkip to conversation

connlimit match fixes for >= 2.6.10

PATCH In review Phil Oester 2 participants 2 messages Feb 12, 2005 – Feb 14, 2005 netfilter

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 connlimit match fixes for >= 2.6.10

Below updates connlimit for:

1) removal of nf_ct_info
2) removal of ctrack

Fixes bugzilla #'s 268 and 286.

Phil

connlimit/linux-2.6/net/ipv4/netfilter/ipt_connlimit.c +8 -5
diff -ru pom-orig/connlimit/linux-2.6/net/ipv4/netfilter/ipt_connlimit.c pom-new/connlimit/linux-2.6/net/ipv4/netfilter/ipt_connlimit.c
@@ -58,6 +58,7 @@
 	int addit = 1, matches = 0;
 	struct ip_conntrack_tuple tuple;
 	struct ip_conntrack_tuple_hash *found;
+	struct ip_conntrack *foundct = NULL;
 	struct ipt_connlimit_conn *conn;
 	struct list_head *hash,*lh;
 
@@ -69,9 +70,11 @@
 	for (lh = hash->next; lh != hash; lh = lh->next) {
 		conn = list_entry(lh,struct ipt_connlimit_conn,list);
 		found = ip_conntrack_find_get(&conn->tuple,ct);
+		if (found)
+			foundct = tuplehash_to_ctrack(found);
 		if (0 == memcmp(&conn->tuple,&tuple,sizeof(tuple)) &&
 		    found != NULL &&
-		    found->ctrack->proto.tcp.state != TCP_CONNTRACK_TIME_WAIT) {
+		    foundct->proto.tcp.state != TCP_CONNTRACK_TIME_WAIT) {
 			/* Just to be sure we have it only once in the list.
 			   We should'nt see tuples twice unless someone hooks this
 			   into a table without "-p tcp --syn" */
@@ -82,7 +85,7 @@
 		       ipt_iphash(addr & mask),
 		       NIPQUAD(conn->tuple.src.ip), ntohs(conn->tuple.src.u.tcp.port),
 		       NIPQUAD(conn->tuple.dst.ip), ntohs(conn->tuple.dst.u.tcp.port),
-		       (NULL != found) ? tcp[found->ctrack->proto.tcp.state] : "gone");
+		       (NULL != found) ? tcp[foundct->proto.tcp.state] : "gone");
 #endif
 		if (NULL == found) {
 			/* this one is gone */
@@ -91,20 +94,20 @@
 			kfree(conn);
 			continue;
 		}
-		if (found->ctrack->proto.tcp.state == TCP_CONNTRACK_TIME_WAIT) {
+		if (foundct->proto.tcp.state == TCP_CONNTRACK_TIME_WAIT) {
 			/* we don't care about connections which are
 			   closed already -> ditch it */
 			lh = lh->prev;
 			list_del(lh->next);
 			kfree(conn);
-			nf_conntrack_put(&found->ctrack->infos[0]);
+			nf_conntrack_put(&foundct->ct_general);
 			continue;
 		}
 		if ((addr & mask) == (conn->tuple.src.ip & mask)) {
 			/* same source IP address -> be counted! */
 			matches++;
 		}
-		nf_conntrack_put(&found->ctrack->infos[0]);
+		nf_conntrack_put(&foundct->ct_general);
 	}
 	if (addit) {
 		/* save the new connection in our list */
#1 lore raw
Quoted 6 lines · Phil Oester

Below updates connlimit for:

1) removal of nf_ct_info
2) removal of ctrack

Fixes bugzilla #'s 268 and 286.

Ok, applied to the pom-ng 2.6 branch . Can you close Bugzilla referring to Rev. 3688 (http://svn.netfilter.org/cgi-bin/viewcvs.cgi/branches/patch-o-matic-ng/linux-2.6.11/?rev=3688)

Thanks!

Signature · 5 lines
- Harald Welte <[email protected]>             http://www.netfilter.org/
============================================================================
  "Fragmentation is like classful addressing -- an interesting early
   architectural error that shows how much experimentation was going
   on while IP was being designed."                    -- Paul Vixie